Sharing R&D Risk in Healthcare via FDA Hedges∗
Adam Jørring,1 Andrew W. Lo,2 Tomas J. Philipson,3
Manita Singh,4 and Richard T. Thakor5
This Draft: 28 March 2017
Abstract
The high cost of capital for firms conducting medical research and development (R&D) has been partly attributed to the government risk facing investors in medical innovation. This risk slows down medical innovation because investors must be compensated for it. We analyze new and simple financial instruments, Food and Drug Administration (FDA) hedges, to allow medical R&D investors to better share the pipeline risk associated with FDA approval with broader capital markets. Using historical FDA approval data, we discuss the pricing of FDA hedges and mechanisms under which they can be traded and estimate issuer returns from offering them. Using various unique data sources, we find that FDA approval risk has a low correlation across drug classes as well as with other assets and the overall market. We argue that this zero-beta property of scientific FDA risk could be a main source of gains from trade between issuers of FDA hedges looking for diversified investments and developers looking to offload the FDA approval risk. We offer proof of concept of the feasibility of trading this type of pipeline risk by examining related securities issued around mergers and acquisitions activity in the drug industry. Overall, our argument is that, by allowing better risk sharing between those investing in medical innovation and capital markets more generally, FDA hedges could ultimately spur medical innovation and improve the health of patients. Keywords: Healthcare Finance, R&D Investments, Drug Development, FDA Approval, Idiosyncratic Risk, Risk sharing, Hedging JEL Classification: G11, G12, G13, G22, G23, G31, G32, I18, K23, L65, O32
∗ We would like to thank Frederico Belo, Mark Egan, Ralph Koijen, Colin Ward, and seminar participants at the Milken Institute for helpful comments and discussions. Any errors are our own. Research support from the MIT Laboratory for Financial Engineering and the University of Chicago Becker Friedman Institute is gratefully acknowledged. The views and opinions expressed in this article are those of the authors only and do not necessarily represent the views and opinions of any other organizations, any of their affiliates or employees, or any of the individuals acknowledged above. 1 University of Chicago, Booth School of Business 2 MIT Sloan School of Management, CSAIL, and NBER 3 University of Chicago, Harris School of Public Policy and NBER 4 Goldman Sachs and MIT Laboratory for Financial Engineering 5 University of Minnesota, Carlson School of Management
15 March 2017 FDA Hedges Page 2 of 41
1. Introduction
Biopharmaceutical and medical device companies typically invest very large amounts of
money in order to develop a treatment. For example, recent estimates suggest that the cost
of developing a single new drug in the biopharmaceutical sector is $2.6 billion (DiMasi,
Grabowski, & Hansen, 2014). However, this development process is also quite risky, not
only due to the inherent scientific risk of developing new compounds for humans, but due
to the risk involved in regulatory approval processes as well. Companies undertaking drug
development in the U.S. are subject to the Food and Drug Administration’s (FDA) approval
process. As a result, given the high cost of development, failure due to either scientific or
regulatory risk during any part of the FDA approval process can have a highly adverse
impact on a company undertaking drug research. Significantly, this risk is borne only by
those investing in the particular treatment under consideration by the FDA and cannot
easily be shared by other investors in the general capital market. Some have argued more
generally that government uncertainty ultimately slows down medical innovation and
hurts future patients (Koijen, Philipson, & Uhlig, 2016).
To overcome this problem, Philipson (2015a,b) suggested financial instruments that
allow those investing in medical innovation to better share scientific and policy-related
development risks with outside investors. We call these types of instruments “FDA hedges.”
In this paper, we discuss the pricing of FDA hedges and mechanisms under which they can
be traded and estimate issuer returns from offering them. In addition, we examine their
risk characteristics and evaluate some unique evidence suggesting a proof of concept that
these risks can be traded in capital markets.
15 March 2017 FDA Hedges Page 3 of 41
If FDA hedges were exchange-traded, there would be direct risk-sharing benefits to
developers who are able to lay off some development risks to other parts of capital
markets. We therefore consider exchange-traded FDA binary options, which pay a fixed
amount of money in the event of a trigger. Binary options are well-known and regularly
traded on various exchanges.6 In the case of an FDA binary option, the triggering event
would be the failure of a specific drug in the FDA approval process. We provide details of
how such a binary option would be priced and use historical data on drug development
success rates by phase and drug type to calculate what the typical price of an FDA binary
option would be for a drug in each therapeutic area.
While the insurance value of FDA hedges is clear to drug developers, the question
remains of what the value will be to the issuer. In the absence of exchange-traded FDA
hedges, we consider over-the-counter (OTC) issuers who might offer a portfolio of FDA
contracts across developers. We simulate the return distribution of such portfolios by
calibrating the data to historical FDA approval rates and estimate the risk/reward profiles
that they imply and how they vary depending on different assumptions related to the
underlying contracts.
A potential advantage of FDA hedges to issuers is that they only depend on pure
scientific risk and do not aim to insure post-approval market risk of a compound. This
makes assessing the risk of these options easier and reduces their correlation to traditional
asset classes such as stocks and bonds. To investigate the correlation patterns of FDA
options, we make use of a novel dataset of project-level time-series estimates of the
6 One difference between bond and FDA option markets in that options do not need to be rated. This facilitates market-making and trading relative to other types of structures.
15 March 2017 FDA Hedges Page 4 of 41
likelihood of eventual FDA approval for thousands of drugs and biologics. We use these
data to construct a panel dataset of the implied prices and returns of FDA options if they
were priced as predicted. We examine the nature of the risk of these synthetic FDA options
and find that the risk is largely idiosyncratic and unrelated to systematic factors. Since the
prices of these hedges are uncorrelated with the broader market or other factors, we argue
that the risk associated with FDA hedges may appeal to investors and issuers interested in
diversification.
Indeed, this zero-beta property could be the main source of gains from trade between
issuers looking for diversified investments and developers looking to offload the approval
risk. It may generally hold if the inherent scientific risk of molecules working in humans
that drive the FDA approval risk were not correlated with other asset classes. A broader
implication of our empirical findings is that the risk of R&D projects is, in general,
idiosyncratic, since the value of FDA options is directly tied to the underlying R&D projects.
To our knowledge, our paper is the first to provide project-level evidence of this point,
which has been posited by a number of papers (e.g. Pastor and Veronesi (2009), Fernandez,
Stein, and Lo (2012), Thakor and Lo (2015)).
Given these risk patterns, we examine how well issuers may be able to hedge the risk of
offering FDA options. We consider the hedge of shorting the stock of the underlying firm
whose drug is going through the FDA approval process and examine the implied value of
such hedges given the prices of the synthetic FDA hedges and the underlying stocks.
Finally, we consider some evidence that FDA risk can be traded, and propose an indirect
measure of the covariance of this type of risk with the broader market. In particular, we
argue that several exchange-traded Contingent Valuation Rights (CVRs) issued in
15 March 2017 FDA Hedges Page 5 of 41
connection with pharmaceutical mergers implicitly offer evidence about the market
acceptance and covariance properties of FDA hedges. The fact that similar risks have been
traded with great liquidity is useful evidence because it negates potential theoretical
arguments that trade may be infeasible due to asymmetric information between
developers and issuers. We consider the price and volume data for these CVRs and examine
their risk. We show that the CVR contracts have no significant exposure to the overall
market and other factors, which provides further evidence that FDA hedges would be
attractive as zero-beta assets to issuers interested in diversification.
Our paper relates most closely to the emerging literature on measuring and analyzing
the economic implications of policy uncertainty on economic activity (Davis (2015)). It also
relates to an emerging literature on the interaction between real- and financial health care
markets and the importance of government risk in slowing down medical innovation
(Koijen, Philipson, & Uhlig, 2016; Thakor, Anaya, Zhang, Vilanilam, & Lo, 2016). We extend
the existing literature by suggesting new financial innovations to try to limit the economic
distortions imposed by policy uncertainty. Our work is also related to the literature that
argues that alternative innovations are needed to mitigate underinvestment in medical
innovation (Fernandez, Stein, & Lo, 2012; Thakor & Lo, 2017).
We start in Section 2 with a discussion of the pricing of FDA binary options and
simulate their prices given historical FDA approval rates and the time they remain in each
FDA phase. In Section 3, we examine the return distributions of pools of FDA hedges
offered by potential over-the-counter issuers. In Section 4, we examine the risk
characteristics of FDA hedges using a panel dataset of FDA approval probabilities and
explore how this risk may be hedged by issuers. In Section 5, we provide the proof of
15 March 2017 FDA Hedges Page 6 of 41
concept of market acceptance of FDA hedges through CVR contracts and analyze the
correlation of the FDA risk with the broader market. We conclude in Section 6 with a
summary of our findings and discuss future research.
2. FDA Binary Options
In this section, we consider exchange-traded FDA binary options and we derive and
calibrate prices for these options in various therapeutic areas.
2.1 Binary FDA Options
Binary options are simple contracts that are currently traded on several exchanges.
Using this concept as a template, we define an FDA binary option as a financial contract
that is sold for a certain price, entitling the holder to be paid a pre-specified amount in the
event that a certain drug fails a given phase of the FDA approval process (or the entire FDA
process), and nothing in the event that it succeeds. An FDA option may be issued at the
start of a given phase for the approval outcome of that phase. Without loss of generality, we
assume it pays one dollar if the drug is not approved, and zero if it is.
Throughout, our pricing formulas will use actual probability estimates to compute
expected values which are then discounted at the risk-free rate. The motivation for this
approach is the fact that the risk associated with FDA approval is unlikely to be correlated
with priced factors such as stock market returns or aggregate consumption. Therefore, the
risk inherent in FDA option payoffs should be solely idiosyncratic, in which case the
equilibrium price should be given by the expected discounted value of the payoff,
15 March 2017 FDA Hedges Page 7 of 41
discounted at the risk-free rate of return. We shall test and confirm this key property
explicitly in Section 4.
Assuming that approval risk is purely idiosyncratic, the price of a binary FDA option is
simply the present value of the probability of non-approval. The two uncertainties is the
outcome of the approval decision itself as well as time when the approval decision is made.
If the approval time is distributed according the frequency f(t), and the probability of non-
approval is p, the price at the start of the phase is given by:7
P =� e-rtpf(t) dt,
where r is the risk-free rate. Clearly, the sooner the decision is made, and the larger the
chance of non-approval, the higher is the price.
2.2 Calibrated Prices of FDA Options
We estimate the prices for binary FDA options using recent evidence on FDA approval
rates. Table 1 below reports the average historical phase failure rates for different disease
groups.8
7 This assumes that there is no correlation between the time of the approval decision and the chance of non-approval. If there is a dependence, we would model the probability as a non-constant function p(t) of time. 8 These failure rates are from Thomas et al. (2016), based upon data from 2006-2015.
15 March 2017 FDA Hedges Page 8 of 41
Table 1: Probabilities of Phase Failure by Disease Group The table shows the average probability of failing each phase of the FDA drug development process, broken down by disease groups. These failure rates are from data from 2006-2015, and are taken from Thomas et. al. (2016).
Probability of Failing Phase Conditional on Reaching It
Disease Group Phase 1 Phase 2 Phase 3
NDA/BLA Approval
Phase
Overall Probability of Failure
Hematology 27% 43% 25% 16% 74% Infectious Disease 31% 57% 27% 11% 81% Ophthalmology 15% 55% 42% 23% 83% Other Disease Groups 33% 60% 30% 12% 84% Metabolic 39% 55% 29% 22% 85% Gastroenterology 24% 64% 39% 8% 85% Allergy 32% 68% 29% 6% 85% Endocrine 41% 60% 35% 14% 87% Respiratory 35% 71% 29% 5% 87% Urology 43% 67% 29% 14% 89% Autoimmune/immunology 34% 68% 38% 14% 89% Neurology 41% 70% 43% 17% 92% Cardiovascular 41% 76% 45% 16% 93% Psychiatry 46% 76% 44% 12% 94% Oncology 37% 75% 60% 18% 95%
Given these probabilities of failure, we calibrate the prices of the FDA binary options
that pay off one million after a given phase if the drug fails that phase. We compute these
prices for contracts structured as single-phase and multiple-phase options. For our
calculations, we assume an annual risk-free interest rate of 1%.
In order to calibrate the timing of FDA decisions (f), we report in Table 2 the average
duration of each phase of the FDA approval process, taken from DiMasi and Grabowski
(2007). The estimates for the phase lengths are different for biotech firms and pharma
firms. We therefore use the average phase length for biotech and pharma firms in our
calculations.
15 March 2017 FDA Hedges Page 9 of 41
Table 2: FDA Approval Process Phase Lengths This table shows the average length of each phase in the FDA approval process for the biotech and pharma sectors. Phase length is in months (years in parentheses). Estimates come from DiMasi and Grabowski (2007).
Average Length of time in months (years)
Sector Phase 1 Phase 2 Phase 3
NDA/BLA Approval
Phase
Total Length of
Time Biotech 19.5 (1.6) 29.3 (2.4) 32.9 (2.7) 16.0 (1.3) 97.7 (8.1) Pharma 12.3 (1.0) 26.0 (2.2) 33.8 (2.8) 18.2 (1.5) 90.3 (7.5) Average 15.9 (1.3) 27.65 (2.3) 33.35 (2.8) 17.10 (1.4) 94.0 (7.8)
Combining the data on approval rates and the timing of FDA decisions, Table 3 reports
the implied prices (if purchased at the beginning of the indicated phase) for single-phase
FDA binary options—options that pay off $1 million if there is failure in the indicated
phase, and nothing otherwise. For the purpose of simplifying our calculations and more
directly conveying the intuition behind the prices of these FDA options, we do not make
distributional assumptions on f and treat the phase length as deterministic by using the
average phase lengths from Table 2 directly in discounting the payoffs of the options. In
other words, the payoff of a single-phase FDA option in Table 3 is given by the following
formula:
P = e-rTpX,
where X is the promised payoff of the option, p is the probability of non-approval, and T is
the average phase length taken from Table 2. We use a risk-free interest rate of 1% in our
calculations. In our simulation results later in the paper, we will make explicit
distributional assumptions on f in our pricing.
15 March 2017 FDA Hedges Page 10 of 41
Table 3: Price of Single-Phase FDA Binary Options The table shows the prices of single-phase FDA binary options, which are issued at the start of each phase and pay off in the event of failure in that phase. Prices are in thousands of dollars.
Price of FDA Option that Pays $1m in a Given Phase ($ thousands)
Disease Group Phase 1 Phase 2 Phase 3 NDA/BLA Approval
Hematology $263 $424 $243 $158 Infectious Disease $301 $560 $266 $111 Ophthalmology $150 $541 $406 $222 Other Disease Groups $329 $589 $296 $114 Metabolic $384 $536 $278 $219 Gastroenterology $241 $628 $383 $76 Allergy $320 $660 $278 $61 Endocrine $406 $585 $340 $138 Respiratory $342 $693 $281 $53 Urology $423 $658 $278 $141 Autoimmune/immunology $338 $667 $368 $138 Neurology $404 $687 $414 $166 Cardiovascular $406 $742 $433 $156 Psychiatry $455 $746 $431 $119 Oncology $367 $737 $583 $174
For example, in Hematology it would cost $243,000 to buy insurance for a $1 million
insurance policy. The prices of the single-phase options correspond directly to the failure
rates in each phase. In particular, the price to purchase an option at the beginning of phase
2 to insure against phase 2 failure is significantly higher than the price to purchase options
at the beginning of the other phases. This reflects the fact that the failure rates in the
development process for the various disease groups are the highest in phase 2. By contrast,
the prices are much lower in the final FDA approval phase, where the failure rates are the
lowest.
We next calculate the prices of multiple-phase FDA binary options, which pay off if
there is failure in any subsequent phase of the FDA process. We discuss the pricing of these
15 March 2017 FDA Hedges Page 11 of 41
options in the Appendix. Table 4 reports the prices of these options if purchased at the
beginning of a given phase, thereby providing insurance against failure in any of the
remaining phases.9
Table 4: The Price of Multiple-Phase FDA Binary Options, for Payoff in each any Subsequent Phase
This table shows the prices of multiple-phase FDA binary options, which are issued at the start of each phase and pay off in the event of failure in any subsequent phase. Prices are in thousands of dollars.
Price of FDA Option that Pays $1m for Failure in Subsequent
Phases ($ thousands)
Disease Group Phase 1 Phase 2 Phase 3 NDA/BLA Approval
Hematology $714 $622 $358 $158 Infectious Disease $784 $704 $344 $111 Ophthalmology $797 $773 $531 $222 Other Disease Groups $812 $734 $373 $114 Metabolic $821 $726 $430 $219 Gastroenterology $821 $778 $428 $76 Allergy $828 $761 $321 $61 Endocrine $843 $753 $428 $138 Respiratory $847 $783 $318 $53 Urology $862 $778 $376 $141 Autoimmune/immunology $862 $807 $451 $138 Neurology $890 $834 $507 $166 Cardiovascular $907 $863 $517 $156 Psychiatry $913 $860 $495 $119 Oncology $921 $893 $650 $174
There are a few noteworthy patterns in the table. First, naturally the price to insure
against any phase rises with non-approval rates. Second, the price of the multiple-phase
option goes down as one advances to subsequent phases, since the conditional probability
9 The details of how these prices are calculated are provided in the Appendix.
15 March 2017 FDA Hedges Page 12 of 41
of the drug failing in the future goes down over time. However, the price that one would
pay for the multiple-phase option only goes down slightly from phase 1 to phase 2,
dropping much more significantly from phase 2 to phase 3, due to the high failure rates in
phase 2. Since the failure rate is much higher in phase 2 relative to all other phases, most of
the cost of the option in phases 1 and 2 will be to insure against failure in phase 2. Once
failure in phase 2 has been averted, the price of the option drops significantly, since failure
is relatively less likely going forward.
3. Risk/Reward Profile of FDA Hedges via OTC Issuers
The insurance value of FDA hedges is clear to drug developers, but the question
remains of what the value will be to those holding the other side of the trade, i.e., the
issuers. In this section, we therefore consider the value to over-the-counter (OTC) issuers
that offer FDA contracts to investors. In order to do so, we simulate the risk and return
distributions of pools of FDA hedges offered by issuers.
3.1 Risk/Reward Profile of Pools of FDA Options
We first empirically investigate the risk and return tradeoff of a pool of FDA option
contracts. We examine a portfolio of N contracts, each linked to a particular FDA
application. If the FDA rejects the application at any t prior to the contract maturity date T,
the issuer pays the insurance buyer $1. The precise timing of the FDA’s approval decision f
is unknown; we model the time until an FDA decision as an exponential distribution with
rate parameter λ. When the FDA reaches a decision before the contract expires, we assume
that the application i is rejected with probability pi, and in our base calculations we assume
15 March 2017 FDA Hedges Page 13 of 41
that there is no correlation between the rejection probabilities of two different
applications, pi and pj. In other words, if each contract represents an FDA option based on
the failure/success of a different drug, the probabilities of failure of each drug are
independent. A priori, this assumption of no correlation across contracts will hold if a
larger probability of one molecule working in humans does not increase the chance of
another. This will likely be the case, except when molecules work within the same
indication or mechanism of action, in which case a correlation may occur.10 In Section 5, we
provide evidence that seems to suggest that the assumption of no correlation between the
contracts would hold in practice.
In our benchmark simulation results, we vary the number of contracts while fixing
other parameters, in order to explore the potential diversification benefits of adding
additional contracts to the issuer’s portfolio. More specifically, we simulate portfolios of N
= 1, N = 10, N = 50, and N = 100 contracts. We assume a contract maturity of T = 5 years,
and pi = 30%. We choose a rate parameter of λ = 1/3 for the time until an FDA decision is
made, in order to match a mean FDA decision time of three years. For robustness, in Table
A1 in the appendix, we provide the portfolio payout distribution characteristics for
alternative choices for the size of the portfolio N, the rejection probability pi, the
correlation across draws ρ, and the arrival rate λ.
We examine the risk-return tradeoff that the issuer faces by calculating the Sharpe
ratios of the portfolios. Consider an issuer who has issued N contracts priced at price $P
with expected payouts X1, …, XN. He invests $NP at the risk-free rate with the return: 10 A correlation would also occur if the FDA decision-making process across molecules is tied together due to regulatory behavior. In the Appendix, we explore how our results are affected when this assumption is relaxed and we allow for correlation between drug applications.
15 March 2017 FDA Hedges Page 14 of 41
R = [NP(1 + r) - ∑Xi]
NP = (1 + r) -
X�P
where X� = ( 1N
)∑Xi. The Sharpe ratio is calculated by dividing the markup by the standard
deviation of the portfolio:
SR = E[R] - r
σ(R) =
P - E[X]σ(X�)
In order to calculate the Sharpe ratios in this setting, we assume contract fees of 2% of the
expected payout of the portfolio, and a risk-free rate equivalent to the current five-year
Treasury yield. We vary the portfolio markup, up to a maximum markup of 50% over
expected portfolio return.
Figure 1 below presents the values of the Sharpe ratio for various values of N as a
function of the portfolio markup. For example, for a portfolio of N = 10 contracts, the
expected payout is estimated to be $2.04, and the standard deviation of the portfolio is
estimated to be 0.449. With a price given by a 35% markup over the expected payout,
contract fees of 2%, and risk-free rate of 1.22%, the Sharpe ratio is calculated to be 1.5546.
As the figure shows, the Sharpe ratio intuitively improves as the markup increases, but an
increase in the number of contracts also consistently improves the Sharpe ratio. Thus, in
the case of independent payoffs amongst the contracts, a larger number of contracts
improve the issuer’s returns. The underlying intuition is the same as that of portfolio
diversification. With any portfolio of assets, introducing uncorrelated assets will reduce the
volatility of the portfolio through diversification.11
11 In Section A.2 of the Appendix, we provide the results for the Sharpe ratios assuming a correlation between the contracts.
15 March 2017 FDA Hedges Page 15 of 41
Figure 1: Sharpe Ratios This figure plots the Sharpe ratios of dealer returns as a function of the portfolio markup % for various values of N, the number of contracts offered in the pool. These calculations assume no correlation between the payouts of the contracts.
3.2 Risk-Return Distributions for Disease Groups
The results above show the risk-return tradeoff faced by issuers for general pools of
FDA option contracts. It is informative to examine in more detail how this tradeoff varies
depending on the particular disease group the FDA Options are based upon, since different
disease groups can have very different success probabilities. Table 5 provides the expected
payout, variance, and Sharpe ratio for a portfolio of FDA options based on a drug project in
each respective disease group (assuming N = 50 contracts in the pool), using the average
probabilities of failure in Phase 3 for each group that were shown in Table 1.
15 March 2017 FDA Hedges Page 16 of 41
Table 5: Expected Payouts of Portfolios of N = 50 Contracts This table provides the simulation results for the mean portfolio payout and variance of payout for different disease groups, assuming N = 50 contracts.
Disease Group Probability of
Approval in Phase III
Expected Payout Variance Sharpe Ratio
Hematology 75% $0.51 0.003 2.39 Infectious Disease 73% 0.50 0.003 2.22 Ophthalmology 58% 0.40 0.003 1.91 Other Disease Groups 70% 0.48 0.003 2.33 Metabolic 71% 0.64 0.004 2.99 Gastroenterology 61% 0.42 0.003 2.08 Allergy 71% 0.48 0.003 3.16 Endocrine 65% 0.44 0.003 2.19 Respiratory 71% 0.48 0.003 2.86 Urology 71% 0.48 0.003 2.36 Autoimmune/immunology 62% 0.42 0.003 2.11 Neurology 57% 0.39 0.003 1.98 Cardiovascular 55% 0.38 0.003 1.94 Psychiatry 56% 0.38 0.003 1.96 Oncology 40% 0.27 0.002 1.51
As can be seen from the table, the portfolio payouts vary between disease groups,
depending on the probability of approval. In particular, the expected payouts by the issuer
are higher if the probability of approval is lower (i.e. the probability that the option will pay
out is higher), with the highest expected payout being in oncology. The variance of the
payouts also increases as the probability of approval decreases. The Sharpe ratio for the
issuer is also generally higher for disease groups with a higher probability of success. For
example, issuers will find that issuing pools of FDA options are more attractive for drugs in
Hematology than for drugs in Oncology, since drugs in Oncology are more likely to fail and
therefore necessitate payouts by the issuer. Overall, the relatively high Sharpe ratios for all
15 March 2017 FDA Hedges Page 17 of 41
of the disease class reinforce the notion that FDA options may be attractive for issuers. For
example, the Sharpe ratio of the S&P 500 SPDR ETF over the past five years was 1.32,
which is substantially lower than the Sharpe ratios presented above.
While the analysis above provides a view into the risk-return tradeoff faced by issuers
of FDA options, the risk that these issuers are exposed to may be further reduced if issuers
are able to hedge the risk of these options. We explore this issue further in the next section.
4. The Risk of FDA Options
In this section, we return to the issue of the nature of the risk of FDA hedges that issuers
face. FDA hedges may have diversification appeal to investors and issuers if the returns to
these securities are uncorrelated with the broader market or other factors, that is, if the
risk of the hedge is idiosyncratic and not systematic. We explore whether this is empirically
the case using a novel dataset of the approval process of drugs. Given this risk, we then
examine the circumstances under which issuers may be able to hedge the risk of FDA
options.
4.1 Dataset Description
We use a novel dataset on the drug approval process from the BioMedTracker Pharma
Intelligence database. This database contains detailed drug trial information for pharma
and biotech companies, including historical approval success rates, development milestone
events, progress updates, and, most importantly, estimates of the likelihood of future FDA
approval for individual drugs in development by each company. The database provides
information on 11,587 drugs across 2,893 different companies. Although the dataset
15 March 2017 FDA Hedges Page 18 of 41
contains information on a handful of development events prior to 2000, it has full coverage
from 2000 to 2016, and we therefore focus on this period for our analysis.
More specifically, we use the reported likelihood of future FDA approval provided by
BioMedTracker in order to construct hypothetical prices for FDA options on a wide variety
of drugs. For each drug and for a given date, BioMedTracker provides an estimate of the
probability that the drug will ultimately be approved by the FDA. These probabilities are
updated each time there is any announcement or other development-related event related
to the particular drug.12 In order to determine likelihood of approval (LOA) probabilities,
BioMedTracker uses a combination of historical approval rates and analyst adjustments
based on development events. More specifically, when a drug development project is
initially started, BioMedTracker assigns an LOA probability to it based on the historical
approval rates of drugs in the particular disease group that the drug belongs to.
BioMedTracker then adjusts the LOA probability for the drug each time a development
event occurs. If the event conveys no relevant information as to the eventual development
success of the drug, then the LOA is unchanged. However, if the event contains relevant
information (for example, trial results), then the LOA is adjusted either up or down by
BioMedTracker depending on whether the information is positive or negative. The
magnitude of the change in LOA is determined by analysts, who evaluate the information
content of the event and assign a change size based upon pre-specified criteria.
As an example, according to BioMedTracker, an event in Phase III that “[m]et primary
endpoint, but with marginal efficacy or no quantitative details; failed primary endpoint but
12 These include a wide variety of events broadly related to the company and drug under development, including trial results and progress updates, regulatory changes, litigation, and company news.
15 March 2017 FDA Hedges Page 19 of 41
strong potential in subgroup; some concern with efficacy vs. safety balance” will cause an
increase in the LOA between 1% and 5%. In contrast, an event which posted “[m]odest
Phase III results or positive results in non-standard subgroup; met primary endpoint but
concerns over safety profile or study design” causes a decrease in the LOA between 1% and
5%. BioMedTracker provides evidence that its LOA estimates have predictive ability in
terms of the eventual success/failure of the drug under development. More specifically,
BioMedTracker notes that, from 2000-2015, 87% of drugs that were eventually approved
had been classified as having an above-average (relative to the disease group) LOA.
Similarly, 75% of the drugs that eventually moved from Phase II to Phase III from 2000-
2015 had been assigned an above-average LOA. 80% of the drugs that were eventually
suspended during the same period had a below-average LOA.
4.2 Risk Exposure of FDA Options
We use this time series data on probabilities of future approval (LOA) to empirically
verify whether the risk of FDA options is idiosyncratic—and thus related only to scientific
risk—or systematic and related to the broader market or other factors. Specifically, we
construct a time-series of synthetic FDA multi-phase binary option prices using the LOA
probabilities described in the previous section. At any given time t, we set the price Fi(t) of
the synthetic FDA option on a given drug project i which pays off $1 if the project fails as:
Fi(t) = exp(-rt(T - t))�1 - LOAi,t�
where LOAt is the LOA probability at time t, rt is the risk-free interest rate at time t, and T - t
is the expected duration of the contract. For simplicity, we use the expected remaining
development time of the drug as a proxy for the expected duration of the contract. We
15 March 2017 FDA Hedges Page 20 of 41
estimate this using the average development times for each phase from Table 2.13 As
before, we use actual probabilities to compute expectations and then discount the expected
value by the risk-free rate because the risk is assumed to be purely idiosyncratic. We later
provide evidence that justifies this assumption. Using this time-series of constructed prices,
we compute the returns for these synthetic options for all drugs in the BioMedTracker
database. We exclude LOA probabilities that are either 0 (the drug has been suspended) or
1 (the drug has been approved), since there is no future development uncertainty for the
drug at those time points.
With these returns, we run regressions to estimate CAPM and Fama and French (1993)
3-factor betas over the period from 2000-2016, and examine whether these betas are
significant. We run these regressions at the option level, as well as the portfolio level by
combining the options into an equally-weighted portfolio. We first use daily data to
estimate the betas. While using daily data has the potential advantage of increasing the
precision of the beta point estimate, one possible concern with using daily data in this
setting is that there is typically no information on each drug between event days, and thus
the return for the FDA option will be zero for those days. While the lack of correlation due
to few events may indeed be valuable to an issuer, for robustness we also provide the beta
estimates using monthly data.
Table 6 below provides the results of these factor regressions. As can be seen from the
table, the coefficients (betas) are insignificantly different from zero for the CAPM and
Fama-French factors, using both daily and monthly data as well as running the regressions
at both the option and portfolio levels. Moreover, the intercept (alpha) estimates are also 13 For example, for a contract currently in Phase 3, we set T - t = 4.204 years.
15 March 2017 FDA Hedges Page 21 of 41
insignificant. This provides evidence that the risk of FDA options is idiosyncratic and not
related to systematic factors, and thus issuing them may be valuable for diversification
purposes. At a broader level, since the value of FDA options are directly tied to the
underlying R&D projects, this provides evidence that is consistent with the risk of R&D
projects being idiosyncratic, a point that has been posited by a number of papers (e.g.
Pastor and Veronesi (2009)).
Table 6: Systematic Risk of FDA Options
This table gives the results of CAPM and Fama-French 3-factor regressions of the excess return of FDA options on the market, size, and value factors. Regressions are run at the option level or portfolio level using either daily or monthly return data from 2000 to 2016, as indicated. Robust standard errors are in parentheses, and are clustered by date when run at the option level. * indicates significance at the 10% level, ** indicates significance at the 5% level, and *** indicates significance at the 1% level.
Dependent Variable: Ri,t – rft
(1) (2) (3) (4) (5) (6) (7) (8) (Mkt – rf)t -0.0003 0.010 -0.0007 0.010 -0.059 -0.0003 -0.061 -0.029 (0.0069) (0.008) (0.008) (0.008) (0.051) (0.059) (0.055) (0.059) SMBt -0.0003 0.015 0.074 0.130 (0.012) (0.025) (0.062) (0.091) HMLt 0.002 -0.026 -0.077 -0.102 (0.019) (0.021) (0.076) (0.111) Constant (α) 0.00003 0.00003 0.00003 0.00003 0.0008 0.0001 0.0007 0.0001 (0.00008) (0.0001) (0.00008) (0.0001) (0.0018) (0.0022) (0.0018) (0.0021) Regression Level Option Portfolio Option Portfolio Option Portfolio Option Portfolio
Data Daily Daily Daily Daily Monthly Monthly Monthly Monthly Obs 20,690,864 3,918 20,690,864 3,918 1,008,291 192 1,008,291 192 R2 0.0000 0.0003 0.000 0.0012 0.0003 0.0000 0.0006 0.0460
4.3 A Direct Test of Idiosyncratic Risk
A potential concern with our factor regressions is that the lack of significance of the
factors may be due to how we discount the payoffs of the options. In particular, if the risk of
15 March 2017 FDA Hedges Page 22 of 41
FDA approval is, in fact, not purely idiosyncratic, then our option pricing formula is
incorrect. In such cases, we should be using the stochastic discount factor to compute
option prices, which amounts to discounting option payoffs using risk-neutral probabilities
instead of actual probabilities to compute expectations. It is therefore possible that we do
not find significant correlation with priced factors because we are not properly accounting
for the pricing kernel.
To address this concern, we examine whether the market return has any significant
predictive power on the success or failure of drugs. The idea behind this test is that any
correlation between FDA option returns and factors such as the market should also
manifest itself in whether drugs ultimately succeed or fail (and thus the FDA option expires
worthless or pays off). Since the success or failure is simply a binary outcome, examining
whether the market return is a factor in predicting this outcome is therefore a way to test
the robustness of our results above without having to discount or rely on estimation of the
pricing kernel. Specifically, we run a logit regression at the drug level where the dependent
variable is a binary variable which equals to one if the drug succeeded (passed U.S.
regulatory approval) on the given day and equals to zero if the drug failed (development
suspension) on the given day. We run this success/failure variable on the
contemporaneous market return, as well as the lagged and forward 20-, 60-, and 90-day
cumulative market returns.
The results of these regressions are given below in Table 7. As can be seen from the
table, the market return is insignificant at every horizon, indicating that the market return
does not have predictive power on the success or failure outcomes of drugs. This provides
further evidence that the risk of FDA approval is purely idiosyncratic.
15 March 2017 FDA Hedges Page 23 of 41
Table 7: Drug Success/Failure Outcomes and the Market Return This table gives the results of logit regressions of drug success or failure outcomes on market returns over different time periods. The dependent variable is equal to one if the drug succeeded on the given day and zero if the drug failed on that day. The market returns are cumulative returns between the indicated lagged or forward date and the day t. Regressions are run at the drug level using daily data from 2000 to 2016. Robust standard errors are in parentheses, and are clustered by date. * indicates significance at the 10% level, ** indicates significance at the 5% level, and *** indicates significance at the 1% level.
Dependent Variable: Drug Success/Failure
Market Return Window:
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Contemporaneous, t -5.201 (4.272) Lagged, t – 1 to t 1.656 (3.182) Lagged, t – 20 to t -0.214 (1.066) Lagged, t – 60 to t -0.314 (0.717) Lagged, t – 90 to t -0.626 (0.510) Forward, t to t + 1 -1.765 (3.002) Forward, t to t + 20 0.338 (1.235) Forward, t to t + 60 -0.127 (0.770) Forward, t to t + 90 -0.464 (0.626) Obs 9,678 9,678 9,678 9,678 9,678 9,676 9,628 9,553 9,474 Pseudo-R2 0.0007 0.0001 0.0000 0.0001 0.0008 0.0002 0.0000 0.0000 0.0003
4.4 Hedging the Risk of FDA Options
In this section, we outline the extent to which an issuer of FDA risk can hedge by trading
the stock of the underlying drug developer. The idea is that any significant movements in
the value of the underlying project that an FDA option is based upon will also affect the
15 March 2017 FDA Hedges Page 24 of 41
stock price of the developing firm. To illustrate in a simple manner, consider a single FDA
option that the issuer hedges by shorting the underlying firm. Let the value of the firm be V
before the approval decision is made, and V1 if approved and V0 if not approved. These
approval-contingent values may be written as:
V1 = X1 + A
V0 = X0 + 0
where (X0, X1) are the value of the assets of the firm due to other factors than the drug
under consideration, and A is the value of the drug under consideration conditional on
approval (and thus equal to zero after non-approval). If X0 and X1 differ, there is a
correlation between the approval decision and the value of the firms due to other factors.
Before the approval decision, the value of the firm is:14
V = pV1 + (1 - p)V0
This equation implies that the price increase due to approval is larger when the probability
of non-approval is larger. Likewise, price drops due to non-approval are smaller when the
probability of non-approval is smaller.
Consider when the issuer of the FDA option shorts the underlying developer to hedge
the FDA option. Consider the case when there is independence between the approval
decision and the other factors driving firm value:, X1 = X0. The payoff of the issuer hedge
after non-approval is then:
V - V0 + P - 1
14 This ignores the possibility that the stochastic discount factor may differ across the two approval states.
15 March 2017 FDA Hedges Page 25 of 41
The first term is positive because the firm loses value and the second term is negative
because the payout on the option is larger than the price charged for it. The payoff after
approval is:
V - V1 + P
The first term is negative because the firm gains value, and the second term is positive
because of the revenue from selling the option comes without a payout.
As an illustrative example of how this issuer hedging may work in practice, consider the
case of Poniard Pharmaceuticals, a firm with a lead drug known as Picoplatin under
development, which is designed to tackle platinum resistance in chemotherapy. While
Picoplatin was under development for a number of different indications, one of its main
indications was small cell lung cancer. According to drug trial data from the
BioMedTracker, Picoplatin for small cell lung cancer was in phase III of the FDA approval
process as of late 2009, when it had a probability of eventual FDA approval of 35%.
Suppose that an issuer had sold a multi-phase FDA binary option as of this point in time,
which pays off in the event that the drug fails any subsequent stage of the development
process or is not approved. Ignoring discounting for simplicity, the price of an FDA option
with a $100 face value will be approximately $100 × �1 - 0.35� = $65.
Now, phase III trial data for Picoplatin for small cell lung cancer was released on
11/16/2009, and the results precipitated a drop in the likelihood of approval for the drug
of 20%, from 35% to 15%. Since the drug was then less likely to be approved, this would in
turn imply an increase in the price of the FDA option from $65 to $100 × �1 - 0.15� = $85,
or a return of -30.7% from the perspective of the issuer’s position. However, suppose that
the issuer also had a short position in the underlying Poniard stock. In the 10 days
15 March 2017 FDA Hedges Page 26 of 41
surrounding the trial data release date, Poniard’s stock posted a return of -70.8%, thus
yielding a return of the short position of 70.8%.15 As a result, on a one-for-one basis, the
short position in the stock more than offsets the increased liability from the FDA option
from the perspective of the issuer. A full hedge in this case would therefore involve a
portfolio with a roughly 50% weight in the short stock and a 50% weight in risk-free
assets.
More generally, we can use the time-series of approval probability data as well as stock
return data in order to estimate the optimal number of underlying stocks needed for
issuers to hedge the risk of FDA options. Let F(t) be the price of the FDA option at date t
that is given by our previous formulas. Denote the underlying stock price return by S(t) and
let n be the number of shares of the underlying stock that issuers hold in order to hedge the
FDA option. The optimal number of shares that minimizes the overall variance of the issuer
satisfies the well-known formula:
n* = �𝜎𝜎𝐹𝐹 σS� ρF,S
where σF is the standard deviation of the FDA option price, σS is the standard deviation of
the underlying stock price, and ρF,S is the correlation between the prices of the FDA option
and the underlying stock.
In order to more clearly illustrate how this hedging may work in practice, we obtain the
approval probability data for the 30 companies in the BioMedTracker database with the
15 One could alternatively examine abnormal returns for the stock, i.e. returns that are attributed to the idiosyncratic movement of the stock (related to the stock’s fundamentals) and not to the market or other systematic factors. Doing so by calculating abnormal returns relative to the market factor yields an even larger drop of 74.8%. The very large drop may indicate that investors viewed the disappointing trial results as an indication that Picoplatin may fail some of its trials for other indications. As a result, in this case it is likely that the drug under consideration is correlated with the other assets of the company.
15 March 2017 FDA Hedges Page 27 of 41
lowest market capitalizations, since these companies are likely to have the fewest number
of drugs or indications in development. We then obtain daily stock price data for these
companies. We eliminate companies for which there are either no drug trial events or for
which there is an insufficient amount of drug trial or stock data. This leaves 19 companies
for which we run our estimation results for.
Using the time-series data on changes in approval probabilities for different drugs to
estimate the prices of multiple-phase FDA binary options written on those drugs, as well as
stock price data for the underlying company stocks, below we estimate the parameters
needed to determine the optimal hedge and the amount of variance reduction it implies for
different drugs. The prices of the FDA options are calculated in the way described in
Section 4.2. Table 8 below presents the optimal hedge for various drugs. The first three
columns correspond to the three parameters above, and the fourth column to the optimal
number of shorted stocks. The fifth column calculates the reduction in variance enabled by
optimal hedging.16
16 Variances and correlations are calculated based on the sample period for which there is data for each drug. For simplicity, we assume a risk-free interest rate of 0 and we ignore the fact that the timing of the FDA approval decision is uncertain. Accounting for this uncertainty will require additional distributional assumptions.
15 March 2017 FDA Hedges Page 28 of 41
Table 8: Optimal Issuer Hedges for FDA Options on Different Drugs
This table gives, for various drugs, the standard deviation of the price of an FDA binary option σF based upon that drug, the standard deviation of the researching company’s stock price σS, the correlation between these prices ρF,S, the optimal number of underlying stocks to purchase n* in order to hedge the option risk, and the reduction in variance implied by the hedge. Company Name Drug σF σS ρF,S n*
Variance Reduction
Acusphere Inc. AI-128 for Asthma 14.13 26.78 -0.42 -0.22 17% Acusphere Inc. CEP-33222 for Breast Cancer 15.02 26.78 -0.37 -0.21 13% Advanced Life Sciences Holdings ALS-357 for Melanoma 2.96 38.63 -0.54 -0.04 10% Advanced Life Sciences Holdings Restanza for Respiratory Tract Infections 4.07 38.63 -0.93 -0.10 82% ARYx Therapeutics ATI-9242 for Schizophrenia 4.32 2.27 -0.74 -1.42 53% ARYx Therapeutics Naronapride for Chronic Idiopathic Constipation 12.29 2.27 -0.84 -4.53 70% ARYx Therapeutics Naronapride for Gastroesophageal Reflux Disease 12.23 2.27 -0.84 -4.51 69% Bone Medical Ltd Capsitonin for Osteoporosis / Osteopenia 2.75 84.04 -0.61 -0.02 4% Boston Therapeutics BTI-320 for Diabetes Mellitus, Type II 0.63 84.04 -0.24 0.00 0% Taxus Cardium Generx for Angina 0.72 84.04 -0.25 0.00 0% diaDexus AIDSVAX for HIV Prevention 1.70 0.34 0.00 -0.02 1% diaDexus PreviThrax for Anthrax Infection (Antibacterial) 10.01 20.40 -0.81 -0.40 65% Entia Biosciences ErgoD2 for Renal Disease / Renal Failure 5.10 20.40 -0.73 -0.18 53% MultiCell Technologies MCT-125 for Multiple Sclerosis (MS) 0.27 108.04 0.57 0.00 8% Neuro-Hitech Huperzine A for Alzheimer's Disease (AD) 9.48 108.04 -0.51 -0.04 23% Neurobiological Technologies Xerecept for Cerebral Edema 1.13 0.37 0.24 0.75 0% Nuo Therapeutics ALD-201 for Coronary Artery Disease 11.87 0.65 -0.76 -13.86 20% Nuo Therapeutics ALD-401 for Ischemic Stroke 10.21 2.74 -0.94 -3.51 88% Nuo Therapeutics ALD-451 for Brain Cancer 4.54 1.03 -0.69 -3.02 41% Ore Pharmaceutical Holdings ORE10002 for Inflammatory Disorders 1.03 1.10 0.13 0.13 1% Ore Pharmaceutical Holdings ORE1001 for Ulcerative Colitis (UC) 0.31 1.10 0.01 0.00 0% OncoVista Innovative Therapies OVI-237 for Breast Cancer 0.54 1.10 -0.10 -0.05 0% OncoVista Innovative Therapies OVI-237 for Gastric Cancer 7.01 0.37 -0.80 -15.10 64% OncoVista Innovative Therapies P-AAT for Acute Coronary Syndrome (ACS) 0.54 10.67 0.29 0.01 0% OncoVista Innovative Therapies P-AAT for Diabetes Mellitus, Type I 2.09 10.67 0.50 0.10 1% Poniard Pharmaceuticals Picoplatin for Colorectal Cancer (CRC) 7.39 0.55 -0.83 -11.07 60% Poniard Pharmaceuticals Picoplatin for Ovarian Cancer 8.50 0.55 -0.78 -11.96 45% Poniard Pharmaceuticals Picoplatin for Prostate Cancer 0.84 0.55 0.20 0.31 1% Poniard Pharmaceuticals Picoplatin for Small Cell Lung Cancer (SCLC) 10.46 1094.51 -0.71 -0.01 8% Poniard Pharmaceuticals Skeletal Targeted Radiotherapy for Breast Cancer 13.53 1094.51 -0.81 -0.01 11%
Poniard Pharmaceuticals Skeletal Targeted Radiotherapy for Multiple Myeloma 13.30 1094.51 -0.35 0.00 5%
Stromacel UMK-121 for Liver Failure / Cirrhosis 13.44 1094.51 -0.81 -0.01 11% Proteo Elafin for Coronary Artery Bypass Graft (CABG) 12.66 1094.51 -0.74 -0.01 12% Rock Creek Pharmaceuticals Anatabine citrate for Alzheimer's Disease (AD) 0.98 1094.51 0.78 0.00 12%
15 March 2017 FDA Hedges Page 29 of 41
Rock Creek Pharmaceuticals Anatabine citrate for Autoimmune Disorders 2.66 1094.51 0.52 0.00 6% Rock Creek Pharmaceuticals Anatabine citrate for Multiple Sclerosis (MS) 0.67 352.99 0.10 0.00 0% Rock Creek Pharmaceuticals Anatabine citrate for Traumatic Brain Injury (TBI) 2.19 2.39 -0.21 -0.19 1% VioQuest Pharmaceuticals Lenocta for Anti-Parasitic and Anti-Protozoal 0.87 2.39 -0.29 -0.11 2% VioQuest Pharmaceuticals Lenocta for Solid Tumors 0.77 2.39 -0.29 -0.09 2% VioQuest Pharmaceuticals VQD-002 for Multiple Myeloma (MM) 14.63 2.39 -0.31 -1.88 3% VioQuest Pharmaceuticals VQD-002 for Solid Tumors 0.59 41.71 0.10 0.00 1%
However, in a number of the other cases, the variance reduction is low—on the
magnitude of 5% or less. There are a few reasons for this. One reason is that, for some drug
indications, there are only a few dates with any news, and moreover there is no change in
the probability of success for many of these dates. Because of this, the price of the FDA
option will remain constant (ignoring discounting) for many dates, and the variance of the
FDA option will be small because the price only changes when there is an event. This may
lead to imprecise inputs into the optimal hedge calculation, and therefore a low variance
reduction. The second reason is that certain drugs or indications make up a relatively small
proportion of the value of a company’s overall drug portfolio. For example, a company may
test a certain compound for efficacy in treatment areas that are different from the drug’s
primary target, and expect a very low likelihood of success. The company’s overall value
will therefore relatively unaffected by clinical news about this indication. As a result, for
these particular types of drugs or indications in development, the underlying stock of the
company may not offer an ideal hedge against an FDA option issued on that drug. But as
noted, for other drugs/indications which make up a substantial portion of the company’s
portfolio, the reduction in variance can be substantial for the issuer.
15 March 2017 FDA Hedges Page 30 of 41
6. Proof of Concept via CVR Contracts
There are several potential theoretical arguments against the liquidity of FDA hedges.
For example, one may argue that trading FDA hedges is infeasible due to asymmetric
information between sellers and buyers, preventing the market from functioning.
Because of these potential objections, in this section we discuss an interesting traded
instrument that provides a “proof of concept” of the liquidity in markets trading FDA risks.
It is similar in many respects to FDA hedges, and the instrument is liquid and follows
predicted pricing and volume patterns. This instrument is a particular version of an
exchange traded contingent valuation rights (CVR) issued in mergers and acquisitions
(M&A) deals, which pays investors pre-specified amounts when certain milestones are met
as part of a M&A deal structure. As these milestones many times include FDA approval
decisions, these traded contracts contain implicit FDA options.
However, one proviso should be kept in mind. Almost all current biopharma CVR’s are
“impure” with respect to FDA approval decisions, as they often include non-FDA related
milestones in addition to FDA approvals. For example, these milestones may include sales
or marketing targets. Due to these additional non-FDA milestones, the daily price
movements of the CVR’s may be driven by other factors not related to FDA approval.
However, this also suggests that the CVR itself is not an adequate hedge against FDA
approval risk, and thus there is need for more pure FDA hedges.
6.1 Contingent Valuation Rights with FDA Options
The contingent valuation right (CVR) is a shareholder right, often given to the selling
shareholders during a merger or an acquisition, which gives the holder a cash payment if
15 March 2017 FDA Hedges Page 31 of 41
certain milestones are achieved. Just as listed companies can be traded on the NYSE or
NASDAQ, CVRs can also be traded on these exchanges. An example of a CVR that was traded
on the NASDAQ is the CVR issued by Celgene on its acquisition of Abraxis. Celgene issued
the Celgene CVR contract, with the holder of the contract entitled to milestone and sales
payments. For the milestone payments, the holder of the CVR was entitled to a fixed sum of
money ($250 million divided by the number of CVRs outstanding) upon FDA approval of
the drug Abraxane for use in the treatment of non-small cell lung cancer by a certain date.
In addition, the holder of the CVR was entitled to another sum of money ($400 million
divided by the total number of outstanding CVR contracts) if the drug Abraxane achieved
FDA approval for use in the treatment of pancreatic cancer. These milestone payments can
be viewed as binary FDA options.
Figure 2 below shows the volume data of the Celgene CVR contract, while Figure 3
shows the price data. In both figures, the top graphs show the volume and price of the CVR
contract, while the bottom graphs show the volume and price normalized as a percentage
of the underlying Celgene stock volume/price. Notice the jump in price around October
2012, when the FDA approved Abraxane for non-small cell lung cancer, and similarly in
November, after a trial that showed promise for pancreatic cancer.
Even though the price of this CVR, by and large, has followed the FDA’s decisions, it is
still an “impure” FDA hedge. For example, it is an unsecured obligation of Celgene, and is
junior to all other claims. It is also callable by Celgene, so there is optionality embedded
into it. The CVR also has sales target payments in addition to milestone payments, which
may in turn carry with it additional risk correlated to the overall market but not FDA risk.
These additional features generate price movements that are orthogonal to any change in
15 March 2017 FDA Hedges Page 32 of 41
the probability of FDA approval, and thus counteract the ability of the contract to act as a
hedge against FDA risk.17
Figure 2: Celgene CVR Traded Volume This figure plots the daily trading volume of the Celgene CVR contract, CELGZ, in number of shares (top figure) and as a percentage of the number of shares traded in the underlying Celgene stock (bottom figure).
17 For this particular CVR, there were also mechanical price changes, such as a large price drop occurring in October 2013 due to the price going ex-dividend.
0.0%5.0%
10.0%15.0%20.0%
25.0%
30.0%
35.0%
40.0%
Oct
-10
Jan-
11
Apr-
11
Jul-1
1
Oct
-11
Jan-
12
Apr-
12
Jul-1
2
Oct
-12
Jan-
13
Apr-
13
Jul-1
3
Oct
-13
Jan-
14
Apr-
14
Jul-1
4
Oct
-14
Jan-
15
Apr-
15
Jul-1
5
Oct
-15
CELGZ Daily Volume, as % of Celgene Volume
15 March 2017 FDA Hedges Page 33 of 41
Figure 3: Celgene CVR Stock Price This figure plots the stock price of the Celgene CVR contract, CELGZ, per share (top figure) and as a percentage of the stock price of the underlying Celgene stock (bottom figure).
An additional example is the CVR issued by AstraZeneca after its acquisition of Omthera
Pharmaceuticals, Inc. in May 2013. This CVR ensured a payment for shareholders of $1.18
per share, provided that specific FDA approvals for investigational cholesterol drug
$0.00
$1.00
$2.00
$3.00
$4.00
$5.00
$6.00
$7.00
$8.00
$9.00
$10.00O
ct-1
0
Jan-
11
Apr-
11
Jul-1
1
Oct
-11
Jan-
12
Apr-
12
Jul-1
2
Oct
-12
Jan-
13
Apr-
13
Jul-1
3
Oct
-13
Jan-
14
Apr-
14
Jul-1
4
Oct
-14
Jan-
15
Apr-
15
Jul-1
5
Oct
-15
Celgene Corporation (CELGZ) - Stock Price
0.0%
2.0%
4.0%
6.0%
8.0%
10.0%
12.0%
14.0%
16.0%
18.0%
20.0%
Oct
-10
Jan-
11
Apr-
11
Jul-1
1
Oct
-11
Jan-
12
Apr-
12
Jul-1
2
Oct
-12
Jan-
13
Apr-
13
Jul-1
3
Oct
-13
Jan-
14
Apr-
14
Jul-1
4
Oct
-14
Jan-
15
Apr-
15
Jul-1
5
Oct
-15
CELGZ Price, as % of Celgene Price
15 March 2017 FDA Hedges Page 34 of 41
Epanova were received by July 31, 2014 and an exclusivity determination was received by
September 30, 2014. An additional payment of $3.52 per share is to be paid if additional
pre-specified FDA regulatory approvals are received by March 31, 2016.
6.2 Correlations and Betas for Contingent Valuation Rights
In Section 5.3, we showed that the risk in synthetic FDA hedges was idiosyncratic. We
now explore whether this is also the case for CVR contracts that are actually traded. Below
in Table 9, we report the CAPM and Fama-French betas of three CVR contracts—Celgene
(CELGZ), Sanofi (GCVRZ), and Wright Medical Group (WMGIZ). We calculate these betas
using both daily and monthly data, in order to ensure that the results are not due simply to
a small time-series sample size. In general, the betas of the contracts are insignificant, even
with features such as sales targets that may include some systematic risk.
For the Celgene CVR contract (Panel A), the market betas (columns (1) and (3)) are
insignificant using both daily and monthly data. When incorporating the Fama-French
factors, the market beta becomes negative and significant using daily data, but not when
using monthly data. Thus there is weak evidence that the Celgene CVR carries some
(negative) market risk. The betas of the Sanofi CVR contract (Panel B) are all insignificant
using both daily and monthly data. Finally, the betas of the Wright Medical Group CVR
contract (Panel C) are all insignificant when using daily data; when using monthly data, the
HML beta becomes significant. However, there are only 37 months of data available for the
WMGIZ contract, and thus the significant beta in column (4) may be an artifact of the small
sample size. Overall, the regression results show that the betas of the CVR contracts are
largely insignificant, which provides additional evidence that FDA hedges are also likely to
15 March 2017 FDA Hedges Page 35 of 41
be uncorrelated with the market if traded in the market, and thus may have diversification
appeal to investors.
Table 9: CVR Factor Regressions This table provides CAPM and Fama-French 3-factor regressions of the excess return of CVR contracts on the market, size, and value factors. Regressions are run using either daily or monthly return data for the Celgene-Abraxane CVR contract (CELGZ) in Panel A, the Sanofi CVR contract (GCVRZ) in Panel B, and the Wright Medical Group CVR contract (WMGIZ) in Panel C. Standard errors are in parentheses. All regressions include a constant term (not reported). * indicates significance at the 10% level, ** indicates significance at the 5% level, and *** indicates significance at the 1% level.
Panel A: CELGZ Contract Dependent Variable: Ri,t – rft
(1) (2) (3) (4) (Mkt – rf)t -0.209 -0.282* 0.808 0.801
(0.133) (0.145) (0.673) (0.735) SMBt 0.350 -0.092
(0.281) (1.224) HMLt 0.154 0.388
(0.307) (1.367)
Data Daily Daily Monthly Monthly Obs 1,379 1,379 66 66 R2 0.002 0.003 0.022 0.023
Panel B: GCVRZ Contract Dependent Variable: Ri,t – rft
(1) (2) (3) (4) (Mkt – rf)t -0.330 -0.270 -0.283 -0.568
(0.220) (0.238) (0.820) (0.888) SMBt -0.345 1.068
(0.468) (1.534) HMLt 0.020 1.582
(0.508) (1.677)
Data Daily Daily Monthly Monthly Obs 1,257 1,257 61 61 R2 0.002 0.002 0.002 0.026
15 March 2017 FDA Hedges Page 36 of 41
Panel C: WMGIZ Contract Dependent Variable: Ri,t – rft
(1) (2) (3) (4) (Mkt – rf)t 0.757 0.771 0.332 0.386
(0.786) (0.798) (1.720) (1.673) SMBt -0.205 0.206
(1.400) (2.305) HMLt -0.186 6.723**
(1.591) (2.800)
Data Daily Daily Monthly Monthly Obs 774 774 37 37 R2 0.001 0.001 0.001 0.150
While the betas of the CVR contracts generally are not significantly different from zero,
it is possible that some other type of risk is common to all these contracts. For example,
there may be a systematic factor other than the market or Fama-French factors that affect
the prices and returns of these contracts. One possibility is regulatory risk, affecting
multiple drugs simultaneously (Koijen, Philipson, & Uhlig, 2016). Another possibility is that
CVR contracts may be based on companies working in similar therapeutic areas, in which
case the success of a drug specific to one company may be correlated with the success of a
similar drug under development by another company.
To explore these possibilities, we examine the correlations of the daily and monthly
returns for the CVR contracts. This correlation matrix is shown in Table 10 below. The table
shows that the correlations between the different contracts are very low and insignificantly
different from zero, suggesting that there is no other common factor is that driving the
15 March 2017 FDA Hedges Page 37 of 41
returns of the CVRs. This provides further evidence that the risk embedded in FDA hedges
is likely idiosyncratic, related to the success of the underlying drugs.
Table 10: Correlation matrix of CVR Returns This table provides correlations between daily (Panel A) and monthly (Panel B) stock returns for the Celgene-Abraxane CVR contract (CELGZ), the Wright Medical Group CVR contract (WMGIZ), and the Sanofi CVR contract (GCVRZ). * indicates significance at the 10% level, ** indicates significance at the 5% level, and *** indicates significance at the 1% level.
Panel A: Daily Returns CELGZ GCVRZ Observations
CELGZ 1,379 GCVRZ 0.015 1,257
WMGIZ 0.009 0.001 774
Panel B: Monthly Returns CELGZ GCVRZ Observations
CELGZ 66 GCVRZ -0.134 61
WMGIZ -0.009 0.107 37
The insignificant betas and low correlation between contracts also underscore an
important point related to the appeal of FDA hedges to OTC issuers. In particular, the
Sharpe ratios to OTC issuers of pools of FDA hedges are substantially lower when the
payoffs of the contracts are correlated. These results provide evidence that the assumption
of no correlation between the payoffs of contracts may be more applicable, and that the
higher Sharpe ratios presented in Section 3.1 with uncorrelated contracts applies.
One alternate explanation for the low betas and covariances of these CVR contracts is
that they have a low trading volume, and therefore they have zero covariance with
anything, since the contracts are not traded. However, even if low trading volume is the
15 March 2017 FDA Hedges Page 38 of 41
cause of the low correlation, the correlation is still low which is valuable to issuers. That
reasoning aside, Table 11 below gives the yearly summary statistics for the trading volume
of the three CVR contracts discussed above.
Table 11: CVR Daily Trading Volume Summary Statistics This table provides summary statistics for the daily trading volume for the Celgene-Abraxane CVR contract (CELGZ), the Wright Medical Group CVR contract (WMGIZ), and the Sanofi CVR contract (GCVRZ). All numbers represent the number of shares traded.
Panel A: Celgene CVR (CELGZ)
Mean Std. Dev. p25 Median p75
2015 17,012.6 20,114.9 3,875 10,950 20,850 2014 21,906.0 44,141.3 5,800 11,600 23,600 2013 67,625.4 216,749.9 4,225 18,000 68,075 2012 52,040.8 182,213.2 3,050 14,900 38,750 2011 35,493.7 140,553.7 2,325 8,950 28,000 2010 70,990.2 85,968.6 22,650 49,500 94,500
Panel B: Sanofi CVR (GCVRZ)
Mean Std. Dev. p25 Median p75
2015 664,032.1 2,055,218.4 110,900 235,250 516,275 2014 850,199.2 1,699,940.0 147,225 348,950 777,450 2013 1,177,137.3 4,337,753.0 74,150 237,050 624,700 2012 609,218.0 1,003,581.5 109,400 207,600 588,150 2011 2,321,230.4 4,181,025.7 529,300 1,054,800 2,424,800
Panel C: Wright Medical Group CVR (WMGIZ)
Mean Std. Dev. p25 Median p75
2015 18,037.3 47,647.4 1,100 4,300 17,700 2014 43,925.8 91,923.1 6,900 17,400 47,450 2013 108,033.8 335,577.3 10,900 33,800 88,625
As can be seen from the table, the mean trading volume each year is significant for all of
the contracts. While the volume for CELGZ and WMGIZ are somewhat similar, the trading
15 March 2017 FDA Hedges Page 39 of 41
volume each year for GCVRZ is large, and significantly higher than the other two. This table
shows that there is significant trading volume for the CVR contracts, and thus the
correlations and betas shown above are likely not due to illiquidity of the contracts.
7. Conclusion
The high cost of capital for firms conducting medical R&D has been partly attributed to
the risk of the regulatory approval process that investors must face in medical innovation
(Koijen, Philipson, & Uhlig, 2016). We investigated new financial instruments, FDA hedges,
to allow medical R&D investors to better share the pipeline risk associated with the FDA
approval process with broader capital markets. Using FDA approval data, we discussed
pricing of FDA hedges and mechanisms under which they can be traded and simulated their
risk and return distributions. We then used a novel panel data set of FDA approval
probabilities to empirically explore the nature of the risk inherent in these contracts and
showed how issuers may effectively hedge this risk. We found evidence that the risk
associated with offering FDA hedges faced by an issuer was largely uncorrelated with other
asset classes. Finally, we offered proof of concept that this type of risk can be traded by
examining related securities issued around M&A activity in the drug industry.
We believe the type of analysis conducted in this paper is a first step in demonstrating
that FDA hedges would enable better risk sharing between those investing in medical
innovation and capital markets more generally. By allowing such risk sharing, FDA hedges
would ultimately help accelerate the development of new medical products and improve
the health of countless future patients. Therefore, financial innovations like these may
better spur medical innovation.
15 March 2017 FDA Hedges Page 40 of 41
References
Davis, Steven J. "Regulatory complexity and policy uncertainty: headwinds of our own
making." Becker Friedman Institute for Research in Economics Working Paper (2015). DiMasi, Joseph A., and Grabowski, Henry G. "The cost of biopharmaceutical R&D: is biotech
different?" Managerial and Decision Economics 28, no. 4-5 (2007): 469-479. DiMasi, J. A., Grabowski, H. G., and Hansen, R. W. "Innovation in the pharmaceutical
industry: new estimates of R&D costs." Medford, MA: Tufts Center for the Study of Drug Development (2014).
DiMasi, Joseph A., Hansen, Ronald W., Grabowski, Henry G., and Lasagna, Louis. "Cost of
innovation in the pharmaceutical industry." Journal of Health Economics 10, no. 2 (1991): 107-142.
DiMasi, Joseph A., Reichert, Janice M., Feldman, Lanna, and Malins, Ashley. "Clinical
approval success rates for investigational cancer drugs." Clinical Pharmacology & Therapeutics 94, no. 3 (2013): 329-335.
Fama, Eugene F., and French, Kenneth R. "Common risk factors in the returns on stocks and
bonds." Journal of Financial Economics 33, no. 1 (1993): 3-56. Fernandez, Jose-Maria, Stein, Roger M., and Lo, Andrew W. "Commercializing biomedical
research through securitization techniques." Nature Biotechnology 30, no. 10 (2012): 964-975.
Koijen, R., Philipson, T., and Uhlig, H. (2016), “Financial Health Economics”, Econometrica 84, no 11 :195–242
Philipson, Tomas, “Hedging Pipeline Risk in Pharma: FDA Swaps and Annuities.” Milken
Institute (March 2015a). Philipson, Tomas. “Saving Lives through Financial Innovation: FDA Swaps and Annuities” Forbes, (March 2015b).
Thakor, Richard T., Anaya, Nicholas, Zhang, Yuwei, Vilanilam, Christian, and Lo, Andrew W.
“Financial Risk and Return in the Biotech and Pharmaceutical Industries from 1930 to 2015”. Working Paper, Sloan School of Management, Massachusetts Institute of Technology, 2016.
Thakor, Richard T., and Andrew W. Lo. “Competition and R&D financing decisions: Theory
and evidence from the biopharmaceutical industry.” No. w20903. National Bureau of Economic Research, 2015.
15 March 2017 FDA Hedges Page 41 of 41
Thakor, Richard T., and Lo, Andrew W. “Optimal Financing for R&D-intensive Firms”.
Working Paper, 2017. Thomas, David W., Burns, Justin, Audette, John, Carroll, Adam, Dow-Hygelund, Corey, and
Hay, Michael. “Clinical Development Success Rates”. Biotechnology Innovation Organization (BIO), 2016.
15 March 2017 Appendix: FDA Hedges Page A–1 of 9
Appendix A: Additional Results
A.1 Multiple-Phase Options
An FDA option may be structured to cover multiple phases of approval, so that it pays
off if there is failure in any subsequent phase of the drug development process. As a simple
example, consider the case where there are four discrete dates in the approval process: t =
1 (phase 1), t = 2 (phase 2), t = 3 (phase 3), and t = 4 (final FDA approval of a New Drug
Application or Biologics License Application). In order to demonstrate the concept more
simply, in the following we assume that each phase is the same length of time, thus
removing the uncertainty related to the time when the approval decision is made. As
before, we use actual probabilities to compute expected values which are then discounted
at the risk-free rate due to the idiosyncratic nature of approval risk. If pt is the probability
that the FDA will approve the drug at time t, then the price of the FDA option at t = 3 will
be:
P3 = exp(-rt) �(1 - p4)X�
The option will be priced recursively at each stage. Therefore, the FDA option which has
the payoff indicated by Figure A-1 below, would be priced at the start t = 0 by:
P0 = exp(-4r) �p1p2p3(1 - p4)X� + exp(-3r) �p1p2(1 - p3)X� + exp(-2r) �p1(1 - p2)X�
+ exp(-r) �(1 - p1)X�
To give an example, suppose that a binary option is structured so that it pays off $1,000
whenever the drug fails the approval process. Assume that the riskless interest rate is 1%
per year, and that the probability of success for each phase of the development process is
the same at 60%. Then purchasing this contract at t = 3 will cost
15 March 2017 Appendix: FDA Hedges Page A–2 of 9
exp(-0.01)�(1 - 0.60)×1000� = $396.02. Purchasing this contract at t = 0, however, will cost
$854.10. The high price relative to payoff reflects the fact that the contract offers full
insurance: it will pay off if the drug development fails during any phase. Alternatively, one
could purchase a contract offering insurance against failure in a specific phase, which
would thus be valued at a lower price. This latter contract may be valuable if the risks of
failure for a particular type of drug are concentrated in a specific phase. For example, the
probability of success for respiratory drugs is significantly lower in phase 2 than it is in any
of the other phases of the drug development process (see Thomas et al. (2016)). As a
result, a binary option that pays off in the event of failure only in phase 2 may be
particularly valuable to a company or an investor that is funding such a drug.
Figure A1: Payoff Diagram of a FDA Binary Option at the Start of Multiple Phases This figure shows the payoff structure of a multiple-phase FDA binary option, when viewed at the beginning of the R&D process. In each branch, pt indicates the probability of success.
15 March 2017 Appendix: FDA Hedges Page A–3 of 9
A.2 Correlated Payoff Calculations and Results
In the analysis in Section 3 of the paper, the payoffs of the individual contracts in the
pool are assumed to be independent. However, as discussed previously, it is possible that
there is some correlation between the outcomes of the various contracts. In this section, we
thus examine the results when relaxing the assumption of independent outcomes, and
introduce a correlation of 0.3 between the payouts of the N contracts.
To explore this, we simulate the X1,…,X50 contracts as Bernoulli random variables, and
we allow for pairwise dependence between all contracts by associating each contract with
a random variable Zi that is normally distributed with mean 0 and variance 1. Zi is
associated with Xi as follows:
Xi = � 1 if Zi < αi 0 if Zi ≥ αi
Here, letting Z1,…,Z50 be distributed according to a multivariate standard normal
distribution with covariance matrix ∑ allows the pairwise correlation among X1,…,X50 to
be captured by the pairwise correlation among the Zi's.
Figure A2 presents the Sharpe ratios for various values of N as a function of the
portfolio markup with this correlation assumption. In this case, the Sharpe ratios are lower
than the case with independent contracts. Moreover, the improvement in the Sharpe ratio
is not monotonic as the number of contracts increase. In particular, while there is a large
improvement in the Sharpe ratio from N = 1 to N = 10, the Sharpe ratios are very similar
between N = 50 and N = 100. The correlation between the contracts reduces the Sharpe
ratio because the correlation increases the standard deviation of the portfolio. Since the
standard deviation enters into the denominator of the Sharpe ratio, a larger correlation will
15 March 2017 Appendix: FDA Hedges Page A–4 of 9
cause the Sharpe ratio to decrease. In this case, introducing correlated assets reduces the
diversification of the issuer’s portfolio, thus reducing the Sharpe ratio. This analysis shows
that the benefit to the issuer of holding contracts critically depends on both the number of
contracts, and on the correlation of the payouts between the contracts. However, as we
previously discussed, a substantial correlation between contracts is not likely to hold in
practice.
Figure A2: Sharpe Ratios, Equicorrelated Contracts This figure plots the Sharpe ratios of dealer returns as a function of the portfolio markup % for various values of N, the number of contracts offered in the pool. These calculations assume a correlation of 30% between the payouts of the contracts.
15 March 2017 Appendix: FDA Hedges Page A–5 of 9
A.3 Portfolio Payoff Simulation Results Across Varied Parameters
Table A1 below provides the portfolio payout mean, variance, and standard deviation
when varying the parameters for the number of contracts N, the FDA decision arrive rate λ,
probability of payout p, and correlation between contracts ρ. Table A2 provides the
portfolio payout mean, variance, and standard deviation for various numbers of contracts N
across the different disease groups.
Table A1: Portfolio Distribution Attributes
This table provides the simulation results for the mean portfolio payout, variance of payout, and standard deviation of payout for various numbers of contracts N, arrival rate parameters λ, for various disease groups, and for probability, and varying correlation parameters.
Number of Contracts λ Mean Variance Std Dev
N = 1
0.20 0.15 0.105 0.32 0.25 0.18 0.117 0.34 0.33 0.20 0.132 0.36 0.50 0.24 0.152 0.39 1.00 0.27 0.176 0.42 1.50 0.28 0.185 0.43 2.00 0.29 0.191 0.44
N = 10
0.20 0.16 0.011 0.10 0.25 0.18 0.012 0.11 0.33 0.20 0.013 0.12 0.50 0.24 0.015 0.12 1.00 0.27 0.018 0.13 1.50 0.28 0.019 0.14 2.00 0.29 0.019 0.14
N = 50
0.20 0.16 0.002 0.05 0.25 0.18 0.002 0.05 0.33 0.20 0.003 0.05 0.50 0.24 0.003 0.06 1.00 0.27 0.004 0.06 1.50 0.28 0.004 0.06
15 March 2017 Appendix: FDA Hedges Page A–6 of 9
2.00 0.29 0.004 0.06
N = 100
0.20 0.16 0.001 0.03 0.25 0.18 0.001 0.03 0.33 0.20 0.001 0.04 0.50 0.24 0.002 0.04 1.00 0.27 0.002 0.04 1.50 0.28 0.002 0.04 2.00 0.29 0.002 0.04
Number of Contracts Probability Mean Variance Std Dev
N = 1
p = 0.2 0.14 0.097 0.31 p = 0.3 0.20 0.133 0.36 p = 0.4 0.27 0.158 0.40 p = 0.5 0.34 0.175 0.42 p = 0.6 0.41 0.182 0.43 p = 0.7 0.48 0.180 0.42 p = 0.8 0.54 0.169 0.41
N = 10
p = 0.2 0.14 0.010 0.10 p = 0.3 0.21 0.013 0.12 p = 0.4 0.27 0.016 0.13 p = 0.5 0.34 0.018 0.13 p = 0.6 0.41 0.018 0.14 p = 0.7 0.48 0.018 0.13 p = 0.8 0.55 0.017 0.13
N = 50
p = 0.2 0.14 0.002 0.04 p = 0.3 0.20 0.003 0.05 p = 0.4 0.27 0.003 0.06 p = 0.5 0.34 0.003 0.06 p = 0.6 0.41 0.004 0.06 p = 0.7 0.48 0.004 0.06 p = 0.8 0.55 0.003 0.06
N = 100
p = 0.2 0.14 0.001 0.03 p = 0.3 0.20 0.001 0.04 p = 0.4 0.27 0.002 0.04 p = 0.5 0.34 0.002 0.04 p = 0.6 0.41 0.002 0.04 p = 0.7 0.48 0.002 0.04 p = 0.8 0.55 0.002 0.04
15 March 2017 Appendix: FDA Hedges Page A–7 of 9
Number of Contracts Correlation Mean Variance Std Dev
N = 1
0.00 0.20 0.133 0.36 0.05 0.20 0.133 0.36 0.10 0.21 0.133 0.37 0.15 0.21 0.133 0.37 0.20 0.20 0.132 0.36
N = 10
0.00 0.20 0.013 0.12 0.05 0.20 0.019 0.14 0.10 0.21 0.025 0.16 0.15 0.20 0.031 0.18 0.20 0.20 0.037 0.19
N = 50
0.00 0.20 0.003 0.05 0.05 0.20 0.009 0.10 0.10 0.20 0.016 0.13 0.15 0.21 0.022 0.15 0.20 0.20 0.029 0.17
N = 100
0.00 0.20 0.001 0.04 0.05 0.20 0.008 0.09 0.10 0.20 0.014 0.12 0.15 0.20 0.021 0.15 0.20 0.20 0.028 0.17
Table A2: Portfolio Distribution Attributes Across Disease Groups
This table provides the simulation results for the mean portfolio payout, variance of payout, and standard deviation of payout for various numbers of contracts N, across different disease groups.
Number of Contracts Disease Group Mean Variance Std Dev
N = 1
Hematology 0.51 0.176 0.42 Infectious Diseases 0.50 0.178 0.42 Opthamology 0.40 0.181 0.43 Other Disease Groups 0.48 0.180 0.42 Metabolic 0.64 0.177 0.42 Gastroenterology 0.42 0.182 0.43 Allergy 0.48 0.179 0.42
15 March 2017 Appendix: FDA Hedges Page A–8 of 9
Endocrine 0.44 0.182 0.43 Respiratory 0.49 0.179 0.42 Urology 0.48 0.179 0.42 Autoimmune 0.42 0.182 0.43 Neurology 0.39 0.181 0.43 Cardiovascular 0.37 0.180 0.42 Psychiatry 0.38 0.180 0.43 Oncology 0.27 0.158 0.40
N = 10
Hematology 0.51 0.018 0.13 Infectious Diseases 0.50 0.018 0.13 Opthamology 0.39 0.018 0.14 Other Disease Groups 0.48 0.018 0.13 Metabolic 0.64 0.018 0.13 Gastroenterology 0.42 0.018 0.14 Allergy 0.48 0.018 0.13 Endocrine 0.44 0.018 0.14 Respiratory 0.48 0.018 0.13 Urology 0.48 0.018 0.13 Autoimmune 0.42 0.018 0.14 Neurology 0.39 0.018 0.13 Cardiovascular 0.37 0.018 0.13 Psychiatry 0.38 0.018 0.13 Oncology 0.27 0.016 0.13
N = 50
Hematology 0.51 0.003 0.05 Infectious Diseases 0.50 0.003 0.05 Opthamology 0.40 0.003 0.05 Other Disease Groups 0.48 0.003 0.05 Metabolic 0.64 0.004 0.06 Gastroenterology 0.42 0.003 0.05 Allergy 0.48 0.003 0.05 Endocrine 0.44 0.003 0.05 Respiratory 0.48 0.003 0.05 Urology 0.48 0.003 0.05 Autoimmune 0.42 0.003 0.05 Neurology 0.39 0.003 0.05 Cardiovascular 0.38 0.003 0.05 Psychiatry 0.38 0.003 0.05 Oncology 0.27 0.002 0.04
N = 100 Hematology 0.51 0.002 0.04
15 March 2017 Appendix: FDA Hedges Page A–9 of 9
Infectious Diseases 0.50 0.002 0.04 Opthamology 0.40 0.002 0.04 Other Disease Groups 0.48 0.002 0.04 Metabolic 0.64 0.002 0.04 Gastroenterology 0.42 0.002 0.04 Allergy 0.48 0.002 0.04 Endocrine 0.44 0.002 0.04 Respiratory 0.48 0.002 0.04 Urology 0.48 0.002 0.04 Autoimmune 0.42 0.002 0.04 Neurology 0.39 0.002 0.04 Cardiovascular 0.38 0.002 0.04 Psychiatry 0.38 0.002 0.04 Oncology 0.27 0.002 0.04